Abstract
Purpose: Preclinical studies suggested that metformin use may protect against glaucoma through multiple mechanisms, while observational studies reported mixed results. To clarify this association, we conducted a prospective cohort study, a meta-analysis, and a drug-target Mendelian randomization (MR) analysis.
Design: A prospective cohort study, meta-analysis, and drug-target MR analysis.
Participants: The cohort analysis included 24 206 UK Biobank participants with diabetes who were free of glaucoma at baseline.
Methods: In the UK Biobank, we examined the association between metformin use and incident glaucoma using multivariable Cox regression. A random-effect meta-analysis was performed to synthesize existing evidence on the association between metformin and glaucoma from human studies. For drug-target MR, genetic variants close to 5 metformin targets genes were weighted by their genetic association with glycated hemoglobin A1c (HbA1c) to proxy their therapeutic effects. The genome-wide association study summary data for primary open-angle glaucoma (POAG) was sourced from the International Glaucoma Genetics Consortium, including only individuals of European ancestry. The drug-target MR analysis was performed using the generalized least squares-inverse variance weighted model.
Main Outcome Measures: Incident glaucoma in UK Biobank, pooled effect estimates from meta-analysis, and MR-derived causal estimates for metformin’s effect on POAG.
Results: During a mean follow-up of 12.1 ± 2.8 years, 841 glaucoma cases occurred, with an incident rate of 2.87 per 1000 person-years. Metformin use was associated with a 19% lower glaucoma risk (hazard ratio, 0.81; 95% confidence interval [CI], 0.67-0.97; P = 0.019) in individuals of European ancestry, independent of glycemic control, intraocular pressure, diabetes duration, and other confounders. Random-effect meta-analysis of cohort studies showed a consistent protective effect (pooled risk ratio, 0.81; 95% CI, 0.72-0.93; P = 0.002) of metformin on glaucoma incidence among subjects with diabetes. Genetically proxied effects of 5 metformin targets, scaled to per standard deviation reduction of HbA1c level, were associated with 34% lower glaucoma risk (odds ratio, 0.66; 95% CI, 0.46-0.94; P = 0.022).
Conclusions: Convergent evidence from cohort, meta-analytic, and genetic analyses supports a possible protective association between metformin use and glaucoma among patients with diabetes. These findings support the need for randomized clinical trials to evaluate metformin’s potential as a preventive glaucoma therapy in individuals with diabetes.
Financial Disclosures: The authors have no proprietary or commercial interest in any materials discussed in this article.